Identifying Arctic states vulnerable to rapid ice loss with a convolutional neural network
Abstract
This work investigates the incidence of rapid ice-loss events in the Arctic, in which a substantial fraction of September sea-ice is lost in only a few years. Such events are seen to occur in climate model simulations forced with increasing greenhouse gas concentrations, but not as yet in the relatively short observational record. We train a convolutional neural network (CNN) using six large climate model ensembles to diagnose vulnerability to rapid ice-loss events. The trained model is shown to separate higher and lower risk states in out-of-sample climate models via a leave-one-out cross-validation. The model emphasises the role of persistently thin sea-ice, remaining sea-ice extent and sustained high upper ocean heat content as predictors of rapid ice-loss. The CNN is applied to observational data over the satelite-era showing a signal for rapid ice-loss prior to a drop around 2007 and an increasing risk from around 2015 to present. Such rapid-ice loss events would likely have dramatic consequences for Arctic ecosystems, coastal communities and global shipping.
Related
- A fully convolutional neural network to classify UXOs using multicomponent electromagnetic induction data
- Predicting nitrogen dioxide column density from the International Wealth Index with a convolutional neural network in Africa
- Marine heat waves in vulnerable coral reef regions are skillfully predictable on multi-year timescales
- Arctic climate tipping points under delayed CO2 emissions mitigation
- Reconciling Arctic sea ice roughness across four observing platforms: a resolution transfer function and the stability of spectral shape over six decades
BibTeX
@misc{patterson2026identifying,
title = {{Identifying Arctic states vulnerable to rapid ice loss with a convolutional neural network}},
author = {Matthew Patterson and Ioana Colfescu and Nathan Kirk and Graeme MacGhilchrist},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.16584},
eprint = {2610.16584},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.16584},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Identifying Arctic states vulnerable to rapid ice loss with a convolutional neural network
AU - Matthew Patterson
AU - Ioana Colfescu
AU - Nathan Kirk
AU - Graeme MacGhilchrist
PY - 2026
DA - 2026/10/12
PB - arXiv
DO - 10.48550/arxiv.2610.16584
UR - https://arxiv.org/abs/2610.16584
AB - This work investigates the incidence of rapid ice-loss events in the Arctic, in which a substantial fraction of September sea-ice is lost in only a few years. Such events are seen to occur in climate model simulations forced with increasing greenhouse gas concentrations, but not as yet in the relatively short observational record. We train a convolutional neural network (CNN) using six large climate model ensembles to diagnose vulnerability to rapid ice-loss events. The trained model is shown to separate higher and lower risk states in out-of-sample climate models via a leave-one-out cross-validation. The model emphasises the role of persistently thin sea-ice, remaining sea-ice extent and sustained high upper ocean heat content as predictors of rapid ice-loss. The CNN is applied to observational data over the satelite-era showing a signal for rapid ice-loss prior to a drop around 2007 and an increasing risk from around 2015 to present. Such rapid-ice loss events would likely have dramatic consequences for Arctic ecosystems, coastal communities and global shipping.
ER -